Knockout and Replacement Gene Surgery to Treat Rhodopsin-Mediated Autosomal Dominant Retinitis Pigmentosa

Abstract

Mutations in the rhodopsin (RHO) gene are the predominant causes of autosomal dominant retinitis pigmentosa (adRP). Given the diverse gain-of-function mutations, therapeutic strategies targeting specific sequences face significant challenges. Here, we provide a universal approach to conquer this problem: we have devised a CRISPR-Cas12i-based, mutation-independent gene knockout and replacement compound therapy carried by a dual AAV2/8 system. In this study, we successfully delayed the progression of retinal degeneration in the classic mouse disease model Rho^P23H , and also Rho^P347S , a new native mouse mutation model we developed. Our research expands the horizon of potential options for future treatments of RHO-mediated adRP.

Introduction

Retinitis pigmentosa (RP) is an inherited retinal disease that leads to visual disability and ultimately results in blindness. With a prevalence of ∼1/4,000, RP affects over 1.5 million individuals globally.¹ Initially, the progressive degeneration of rod cells presents as night blindness in patients, subsequently leading to a secondary degeneration of cone cells, ultimately resulting in a complete loss of photosensitivity.² Approximately 60 genes, encompassing over 3,000 mutations, have been identified to be associated with RP.^3,4 Among them, the mutation in rhodopsin (RHO) gene is the most common cause of autosomal dominant retinitis pigmentosa (adRP).⁵

Over 150 types of RHO gene mutations have been identified as causative factors for the disease.⁶ RHO is a G-protein-coupled receptor protein localized in the outer segment of rod cells.⁷ It comprises seven transmembrane domains and plays a crucial role in maintaining vision in low-light conditions.^8,9 As of now, there are no approved drugs that can treat or prevent RHO-mediated adRP.

With in-depth research on the RHO-mediated disease animal models, researchers made a significant discovery: the more allelic genes with RHO point mutations or the higher the expression of incorrect RHO protein, the more severe the disease manifestation.^10
–12 This suggests that the Rho point mutation might be a gain-of-function mutation. For genetic diseases like RHO-mediated adRP, traditional gene replacement therapies are no longer applicable; instead, suppressing the production of toxic proteins becomes crucial. To address this challenge, researchers initially adopted a direct approach. Leveraging the sequence differences between allelic genes caused by point mutations, they used nucleic acid editing technologies such as CRISPR or RNAi to specifically eliminate erroneous DNA or RNA, thereby providing a gene therapy avenue for RHO-mediated adRP.^13

–16

However, this method is limited in that it only targets a single point mutation, which is very unfavorable for future clinical applications. Clearly, for RHO-mediated adRP with over 150 mutations, formulating a universal treatment plan is imperative. This led to the emergence of a knockout and replacement strategy.^17

–21

In the past 2 years, some research teams have been using SpCas9 as a gene editing tool to implement the knockout and replacement treatment strategy. They adopted adeno-associated virus (AAV) as the therapeutic vector and achieved encouraging results in the treatment of the hRho^C110R -transgenic mouse model.²² However, the relatively large size of SpCas9 presents limitations when combined with AAV gene therapy, such as significantly restricting the ability to use other components in the therapeutic vector, preventing us from selecting more appropriate promoters. Furthermore, the potential off-target effects of SpCas9 also pose a safety concern for gene therapy.²³

In previous research, our laboratory developed a new gene-editing tool—Cas12i^HiFi.²⁴ With its smaller size and lower off-target rate, Cas12i^HiFi has shown the potential to be a more superior gene therapy tool. In this study, we used this novel gene editing tool, CRISPR/Cas12i^HiFi, and executed a comprehensive RHO knockout and replacement strategy to treat the Rho^P23H adRP disease model. This method successfully delayed retinal degeneration for as long as 9 months. To validate the universality of our treatment strategy, we also established a new mouse model with a native point mutation in Rho, P347S. In treating this new model, our approach also successfully slowed down the disease progression. In summary, we have successfully developed a universal treatment strategy combining the novel gene-editing tool with dual AAV for RHO-mediated adRP.

Materials and Methods

Animals

The Rho^P23H model mice were brought from Jackson laboratory (Strain: #017628). The C57BL/6, Rho^P23H/+ and Rho^P347S/+ mice were maintained in the animal care facility of the Institute of Zoology, Chinese Academy of Sciences. All animal experiments were performed according to the Guidelines for the Use of Animals in Research issued by the Institute of Zoology, Chinese Academy of Sciences.

Construction of Cas12i^HiFi plasmid and sgRNA

The construction method of related plasmids is mainly to construct or synthesize by GenScript based on previous research results.²⁴ The sequence of Cas12i^HiFi-sgRNA is listed in Supplementary Table S1.

Subretinal injections

Mice were anesthetized by intraperitoneal injection of tribromoethanol (150 mg/kg). Pupils were subsequently dilated with tropicamide (0.5%) and phenylephrine hydrochloride (0.5%). A small hole was made through the cornea limbus at the temporal side using a 30-gauge needle (305106; BD Biosciences) under a stereoscope. A 19 mm 33-gauge blunt-end needle (7803-05; Hamilton) fitted to a Hamilton syringe (7633-01; Hamilton) was then inserted through the tunnel. One microliter of AAV or phosphate-buffered saline (PBS) was then injected slowly into the subretinal space. Fluorescein (100 mg/mL; Alcon Laboratories, Inc.) was included in the suspensions (0.1% by volume). Carbomer Eye Gel (Carbomer, 0.2%; Alcon Laboratories, Inc.) was applied to each eye after the surgery. Eyes with leakage of vector solution or severe injury were excluded from subsequent study.

Production of AAV

All AAV viruses used in this study were produced by PackGene Biotechnology Company, with an initial dosage of 1E10 vg/mL and a total volume of 500 μL. Plasmids required for AAV packaging are synthesized by GenScript.

Editing efficiency detection by T7EI

We extracted DNA from the transfected cells by One Step Mouse Genotyping Kit (#PD101-01; Vazyme, China), following the manufacturer's guidelines. Then we performed polymerase chain reaction (PCR) using specific primers, primers are listed in Supplementary Table S1. Take out 10 μL PCR product and anneal. Then we added the appropriate amount of enzyme and buffer according to the T7EI operating manual (NEB). Finally, through nucleic acid electrophoresis (Bio-Rad), we can roughly understand the different efficiencies of gene editing.

RNA extraction and RT

The transfected cells were sorted for GFP-positive signal by fluorescence-activated cell sorting (FACS). The RNAs of GFP-positive cells were extracted using PureLink RNA Mini Kit (Life). The reverse transcription of RNA was performed using the HiScript III 1st Strand cDNA Synthesis Kit (Vazyme) after removing DNA using the gDNA wiper provided in the kit. The obtained cDNA is stored in a −20°C refrigerator.

Real-time fluorescence quantitative PCR

Take out the processed DNA, prepare the reagents according to the instruction of the AceQ Universal SYBR qPCR Master Mix Kit (Vazyme). Finally, the quantitative PCR (qPCR) experiment is running on the Q6 (Thermo-Fisher). By converting the obtained Ct values, we got the expression of different genes. The primers required for qPCR are shown in Supplementary Table S1.

Western blot

To extract mouse retinal proteins, we removed the mouse eyeballs and peeled off irrelevant tissues such as vitreous body and retinal pigment epithelium under a microscope. Finally, the mouse retinas were soaked in RIPA solution (Beyotime Bio-Technology, China). After lysing on ice for 2 h, add 5 × loading buffer (10 mL; 1.25 mL 0.5 M pH 6.8 Tris-HCl, 2.5 mL glycerin, 2 mL 10% sodium dodecyl sulfate (SDS), 200 μL 0.5% bromophenol blue, 3.55 mL H₂O, and 0.5 mL β-mercaptoethanol) and boil for 5 min. These samples were run on SDS–polyacrylamide gel electrophoresis (SDS-PAGE) with a 5% stacking gel and a 10% separating gel at 100 V for 1 h. This was followed by transferring to a nitrocellulose membrane at 200 mA for 1 h at 4°C.

After blocking the membrane in a TBST buffer supplemented with 3% bovine serum albumin (BSA) for 1 h, it was treated with the primary antibody overnight at 4°C. Following a TBST wash, the membrane was exposed to the secondary antibody for 1 h. Signal detection was accomplished using electrochemiluminescence and films after subsequent washes.

Immunofluorescence

Mouse eyes were collected immediately after killing, fixed in FAS (Servicebio G1109) for 24 h. The eyes were processed stepwise through a dehydration series (75%, 85%, 90%, 95%, 100% ethanol, alcohol benzene, and xylene). Five-micrometer sections were cut with a Leica slicing machine (RM2016; Leica Biosystems, Wetzlar, Germany) after paraffin embedding. The paraffin sections were dewaxed and hydrated, followed by a 45 min blocking with 2.5% BSA and 5% donkey serum in PBS. The sections were incubated overnight at 4°C with primary antibody diluted in 2.5% BSA and 5% donkey serum. After washing three times in PBS, sections were incubated with secondary antibodies and Hoechst 33342 for 1 h. Images were captured by Zeiss LSM 880 confocal laser scanning microscope (Zeiss, Germany). Antibodies used in this study are listed in Supplementary Table S2.

Histological analysis

For histological evaluations, mouse eyes were fixed overnight at room temperature in Modified Davidson's Fluid (with a ratio of formaldehyde:glacial acetic acid:ethanol:H₂O being 6:3:1:10). The eyes then underwent a progressive dehydration process using ethanol concentrations of 70%, 80%, 90%, and finally 100%, before being embedded in paraffin. Retinal sections, aligned with the optic disc on the sagittal plane, were precisely sliced using a Leica RM2235 machine (Leica Biosystems) and then placed on poly-d-lysine-coated slides (Zhong Shan Golding Bridge Biotechnology, Beijing, China). After the dewaxing and rehydration steps, the sections were treated with hematoxylin and eosin (HE) stains following standard protocols. Imaging was conducted using a Leica Aperio VERSA 8 microscope (Leica Biosystems).

Retinal whole-mount

Mouse eyes were enucleated as quickly as possible after killing. Eyes were washed gently and incubated in chilled PBS for 15 min. Eye balls were then fixed in FAS (G1109; Servicebio) for 1 h. The whole retina was harvested and placed on glass slide with photoreceptors layers facing up. The tissue was incubated with Hoechst 33342 for 1 h after washing three times in PBS.

Electroretinography

Signals were recorded using the Visual Electrophysiology Instrument (electroretinogram [ERG]; Optoprobe Science Ltd.). After overnight dark adaption, mice were anaesthetized by intraperitoneal injection of tribromoethanol (150 mg/kg) and pupils were subsequently dilated with tropicamide (0.5%) and phenylephrine hydrochloride (0.5%). Small silver wire loops were placed on each cornea. Both cheeks were injected by silver needles reference electrode, and a ground electrode was placed subcutaneously in the tail. Responses from both eyes were recorded simultaneously.

Retinal dissociation and flow cytometry

Mouse eyes were collected after killing. Retinas were isolated and dissociated using the Papain Dissociation System (LK003150; Worthington). Single cells were stained by propidium iodide staining buffer (BB-4142; Bestbio) to exclude dead cells. Cell suspension was purified and analyzed by FACS (BD Fusion). To minimize the interference of nonphotoreceptor cells in retinal experiments, this study implemented flow cytometry for the preliminary screening of the retinal cell suspension postdigestion. This step significantly enhanced the purity of rod cell samples, thereby providing high-quality specimens for subsequent experimental research. For specific operational details, please refer to the research conducted by Feodorova laboratory.²⁵

Construction of Rho^P347S knock-in mice

Previously, T7-spCas9 and T7-sgRNA vectors were described by our laboratory.²⁶ The T7-sgRNA vector was digested with BsaI (New England Biolabs) and gel purified. SpCas9 mRNA was in vitro transcribed using ARCA mRNA kit (with tailing) (NEB) following the user's guide (New England Biolabs). sgRNAs were in vitro transcribed using HiScribe^® T7 quick high-yield RNA synthesis (NEB). The donor is a 120 nt single-stranded DNA synthesized by Beijing Genomics Institute. The sequence of donor and sgRNA is listed in Supplementary Table S1.

Intracytoplasmic RNA and DNA microinjection

The procedure of RNA microinjection within the cytoplasm was conducted as described in an earlier study.²⁶ In summary, embryos at the one-cell stage were harvested at 0.5 days after mating. Each zygote underwent a cytoplasmic microinjection consisting of 25 ng sgRNA, 100 ng SpCas9 mRNA, and 0.1 nmol ssDNA-donor. Post-microinjection viable embryos were transferred on that very day into the fallopian tubes of ICR pseudopregnant mice. Newborn pups were later delivered through natural labor.

Genomic DNA extraction, PCR, and sequencing for Rho ^P347S mice

We extracted the DNA from mouse tail tips using the One Step Mouse Genotyping Kit (#PD101-01; Vazyme), following the manufacturer's guidelines. We then amplified the Rho^P347S genomic region using both outer and inner primers (Supplementary Table S1). The PCR products generated by the outer primers were subsequently integrated into TA-cloning vectors and introduced into competent Escherichia coli cells. After allowing the cells to culture overnight, we randomly selected colonies for Sanger sequencing.

Quantification and statistical analysis

In the figure legends, we have detailed various statistical metrics, including methods of analysis, significance levels, and n-values. The Prism Software (GraphPad) was used for our statistical evaluations. To check for data normality, we used the Shapiro–Wilk test. If the data adhered to normal distribution, we utilized either one-way ANOVA or two-way ANOVA, depending on the data set. For non-normally distributed data, relevant nonparametric tests such as Kruskal–Wallis or Mann–Whitney U-test were used. We set the threshold for statistical significance at p = 0.05 across all analyses; *p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001. Unless specified otherwise, data representation follows the format: mean ± standard deviation.

Results

Dual AAV-Cas12i^HiFi-based Rho knockout

In our previous research, we have established a highly efficient and specific cellular-level editing system known as CRISPR/Cas12i^HiFi.²⁴ We first tested multiple sgRNAs and evaluated the editing efficiency in the mouse neuroblastoma N2A cell line. Four candidate sgRNAs were selected, through T7 endonuclease I (T7EI) assay and Sanger sequencing (Fig. 1a). Subsequently, four candidate sgRNAs and a scramble gRNA were individually encapsulated into AAV vectors for in vivo validation (Fig. 1b). Here, to enhance the delivery of the system to photoreceptor cells, we used recombinant adeno-associated virus serotype 8 (AAV8), known for its high infectivity for the retinal tissues.

Figure 1.

Design and evaluation of knockout and replacement system. (a) Indel efficiency comparison of different Cas12i^HiFi-sgRNAs on the mouse Rho in mouse N2A cells, determined by T7EI assay and Sanger sequencing. The designed location is focused on the position of exon 1. (b–d) In vivo knockout efficiency of the four sgRNAs. At 14 days of age, WT mice were administered a 5E9 vg dose of AAVs. These AAVs were individually packaging each of the four different sgRNAs as well as a scramble sgRNA. Two-week postinjection, (c) qPCR and (d) western blot experiments were conducted; NC, the protein extract of the N2A cell line. The error bars indicate the SD, the method for significance difference analysis is one-way ANOVA; n = 3. ***p < 0.001. Schematic of (e) knockout dual-AAV system and (h) replacement AAV system. The violet box represents the U6 promoter and P2A sequence is shown in brown box. (f, g) Knockout efficiency of Rho by dose gradient. Expression of Rho in mouse retina 1-month postinjection as assessed by (f) qPCR and (g) western blot. NC, the protein extract of the N2A cell line. The error bars indicate the SD, the method for significance difference analysis is one-way ANOVA; n = 3. ***p < 0.001. (i) RHO restoration through different AAV dosage injection by qPCR. The error bars indicate the SD, the method for significance difference analysis is one-way ANOVA; n = 3. ****p < 0.0001. AAVs, adeno-associated viruses; NC, negative control; qPCR, quantitative polymerase chain reaction; RHO, rhodopsin; SD, standard deviation; WT, wild-type.

We further evaluated the in vivo activity of the system by subretinal injection, a highly effective drug delivery method to photoreceptors. We harvested retinas at 2 weeks postinjection to assess Rho mRNA expression levels and Rho protein content. The results indicate that sgRNA4 significantly reduces the expression level of Rho, demonstrating an excellent in vivo Rho knockout effect (Fig. 1c, d).

Since knockout and replacement need to be accomplished simultaneously,^18
–20 it is essential to ensure that the CRISPR system is only activated when both AAVs coexist in the same target cell. Therefore, we placed the crRNA components and the hRHO gene in one AAV, while the other AAV contains only the rod-specific promoter and the Cas12i^HiFi (Fig. 1e, h). To identify the optimal balance between knockout and replacement, and to achieve the best therapeutic effects, we further conducted in vivo dose-responsive validation. Cas12i^HiFi and sgRNA were packaged in two AAV8 vectors, driven by hRHO promoter and U6 promoter, respectively (Fig. 1e). Preliminary experiments indicated that hRHO promoter can efficiently and specifically drive the expression of target genes in mouse rod cells (Supplementary Fig. S1). Wild-type (WT) mice were injected with AAVs across a dose range of 1E7 to 1E10 vg per eye into the subretinal space.

Results from western blot and qPCR demonstrated that AAV doses higher than 5E8 consistently achieved effective knockdown of Rho (Fig. 1f, g). In the complementary experiment of human RHO replacement, doses of 5E8 vg and above achieved efficient protein restoration (Fig. 1h, i). Based on previous studies, excessively high AAV doses exhibit cytotoxicity.²⁷ Therefore, considering a balance between safety and efficacy, we ultimately chose the dose of 5E8 vg for treatment in the mouse model.

Subretinal delivery of the dual AAV system efficiently coinfects photoreceptor cells

To further investigate the coinfection efficiency of the dual AAV system in photoreceptor cells, we established a dual AAV system, which individually packaging GFP and mCherry. Dual AAV system was injected into the subretinal space of C57BL/6 mice at postnatal day 14 (P14). The majority of the retina was coinfected by both AAVs at 1-month postinjection (Fig. 2a). Hence, to determine the proportion of cells, coinfected by the dual AAVs, the retinas from the injected mice were digested into single cells by papain dissociation system. Subsequently, GFP-positive (GFP⁺) and mCherry-positive (mCherry⁺) cells were sorted by FACS for further analysis (Fig. 2b). Encouragingly, ∼80% of the photoreceptor cells were coinfected by both AAVs (Fig. 2c, d). These results indicate that the dual AAV system we established is capable of achieving extensive coinfection of photoreceptor cells.

Figure 2.

Establishment of a dual AAV subretinal delivery system. (a) Wholemount and sectional analysis of subretinal injection of dual AAV packaging GFP and mCherry. Localization of GFP and mCherry expression within the region enclosed by yellow dashed lines. The white arrowhead indicate cells that clearly exhibit colocalization of red and green fluorescence signals by immunofluorescence staining. (b) Single cell dissociation of retina followed by flow cytometric analysis. Retina was harvested and isolated by papain system. Cells were sorted by FACS into bulk. (c, d) Analysis of GFP and mCherry expression in flow cytometry of cells. The error bars indicate the SD, n = 24. FACE, fluorescence-activated cell sorting.

Long-term preservation of retinal function and structure with dual-AAV strategy (knockout+replacement)

To assess the effectiveness of our therapeutic strategy, we first utilized disease model retained point-mutation P23H in RHO, frequently observed in human RHO diseases.^28,29 Existing literature comprehensively depicts the early stages of rod cell degeneration in Rho^P23H mice,³⁰ followed by secondary cone cell death.³¹ Currently, the P23H model is an excellent model for testing the effectiveness of therapeutic approaches. On P14, therapeutic AAV at a dose of 5E8 vg was delivered subretinally to the right eye of mice. Concurrently, the left eye received an equivalent dose of the GFP vector, serving as a control to account for potential variations in treatment outcomes arising from individual mouse differences. Subsequently, we evaluated the visual function at intervals of 1, 3, 6, and 9 months postinjection (Fig. 3a).

Figure 3.

Evaluation of ERGs and retinal thickness after dual AAV injection on Rho^P23H/+ mice. (a) Design of knockout and replacement therapy strategies. The therapeutic vector was packaged in dual AAV. At P14, two eyes were treated with 5E8 vg therapeutic AAV and AAV-GFP, respectively. The violet box represents the U6 promoter and P2A sequence is shown in brown box. (b) Representative ERG waveforms from Rho^P23H/+ mice postinjection. Dual AAV or AAV-GFP was injected into the subretinal space bilaterally in Rho^P23H/+ mice on P14. Left panel, light intensity = 0.01 cd/(s·m²); right panel, light intensity = 3.0 cd/(s·m²). (c) Quantification of dark-adapted a-wave and b-wave amplitude in Treatment group and AAV-GFP group (Control). The error bars indicate the SD, the method for significance difference analysis is two-way paired ANOVA; *p < 0.05; **p < 0.01; ***p < 0.001; PI-3M group n = 14, PI-6M group n = 17. (d) Immunofluorescence staining of RHO and GFP on retinal sections from Rho^P23H/+ mice after injection. (e) Measurement of ONL cell layer. The error bars indicate the SD. The method for significance difference analysis is two-way paired ANOVA; n = 8. (f) Counting of rod cells per 50 μm. The number of RHO positive cells at random 50 μm sites in ONL was counted and recorded, n = 8. *p < 0.05; **p < 0.01; ***p < 0.001. ERGs, electroretinograms; ns, no significant; ONL, outer nuclear layer.

It is noteworthy that through the analysis of ERG measurements taken at various time points postinjection, both eyes of the mice demonstrated a degree of ERG amplitude attenuation under light intensities of 0.01 and 3 cd/(s·m²). However, we observed a notably more pronounced decline in the ERG of the control group compared to the treatment group. Notably, representative ERG waveforms between the control and treatment groups showed no significant difference at 1-month postinjection. By ninth month postinjection, the ERG waveforms of the control group had flattened, indicating a loss of visual function in the control eyes, while the treatment group maintained relatively good visual acuity (Fig. 3b).

Also, comprehensive analysis of ERG data from numerous mice, under light intensities of 0.1 and 3 cd/(s·m²), revealed no significant differences in ERG amplitude between the treatment and control groups after 1 month postinjection (Supplementary Fig. S2). With the progression of time under a 3.0 cd/(s·m²) light intensity stimulus, the b-wave amplitude of the ERG in the treatment group was significantly higher than that of the control group 3 months postinjection (Fig. 3c), and this difference became even more pronounced 6 months postinjection (Fig. 3c).

The initial site of pathology in RHO-mediated adRP disease affects the rod photoreceptor cells located beneath the retina. As these visual cells undergo apoptosis, there is a corresponding reduction in retinal thickness. We harvested retinal tissues at 3, 6, and 9 months postinjection for analysis. Upon immunofluorescence staining, the retinal architecture in the treatment group (Treatment) was more complete than in the control (AAV-GFP) group (Fig. 3d). This result was further corroborated by comparing the cell layers of the outer nuclear layer across different time points (Fig. 3e) and the number of rod cells per 50 μm (Fig. 3f). Collectively, these results indicate that our intervention to the Rho^P23H/+ effectively mitigated visual degeneration. Furthermore, it underscores the protective efficacy of our dual-AAV therapeutic strategy on photoreceptors.

Derivation of Rho^P347S/+ knock-in mice

The point mutation P347S, recognized as the second major mutation type associated with RHO disease, also plagues human.²⁹ While numerous modified human-P347S transgenic mouse disease models are available,¹⁰ none currently exists the inherent P347S mutation in mice. To demonstrate the universality of our therapeutic strategy, we generated a mouse model for this pathological type, Rho^P347S .

In earlier studies,^32,33 it was demonstrated that targeted precise mutations in DNA sequences can be achieved using the CRISPR-Cas9 system in conjunction with meticulously designed donor-DNA. To facilitate precise gene editing, we selected an sgRNA sequence proximal to the mutation site and carefully crafted a single-stranded donor DNA (Fig. 4a) Then, a mixture of SpCas9 mRNA, sgRNA, and donor DNA (ssDNA) was injected into the zygotes of C57BL/6J mice (Supplementary Fig. S3a). We naturally mated founder-mice and obtained a large number of F1 offsprings. Through subsequent PCR and AflII enzyme digestion analysis, we successfully identified and obtained homozygous knock-in (Homo-) mice and heterozygous knock-in (Heter-) mice (Supplementary Fig. S3b). Finally, the results were confirmed by Sanger sequencing (Fig. 4b).

Figure 4.

Construction and detection of Rho^P347S model mice. (a) Strategy for constructing model mice with the P347S point mutation in the Rho gene. The asterisk (*) represents the position of the P347S point mutation. (b) Validation of Sanger sequencing results. The position of aa347 is inside the purple dashed box. The sequence above represents the Sanger-sequencing result of WT; the sequence below is post-P347S-mutation. (c) Representative ERG waveforms from three different genotypic mice (WT, Rho^P347/+ and Rho^P347S/P347S ), 1.5 month after birth. Left panel, light intensity = 0.01 cd/(s·m²); right panel, light intensity = 3.0 cd/(s·m²). (d) Quantification of dark-adapted a-wave and b-wave amplitude in WT group, Rho^P347S/+ group and Rho^P347S/P347S group, 1.5 month after birth. Each dot represents a biological replicate. The error bars indicate the SD. The number of samples in each group is 8 (n = 8). The method for significant difference analysis is two-way ANOVA; ns: no significant; **p < 0.01; ****p < 0.0001. (e) H&E staining of WT, Heter- (Rho^P347S/+ ) and Homo- (Rho^P347S/P347S ) mice, 1.5 month after birth. Scale bar, 20 μm. (f) Quantitative analysis of dark-adapted a-wave and b-wave amplitude in WT group and Rho^P347S/+ group under progressively increased light intensity stimulation, 2.5 month after birth. Each dot represents a biological replicate. The error bars indicate the SD. The number of samples in each group is 6 (n = 6). Two-way ANOVA; ns: no significant; *p < 0.05; **p < 0.01; ****p < 0.0001. GCL, ganglion cell layer; INL, inner nuclear layer.

To validate the phenotype of this disease model, ERG testing was performed on mice at 1.5 months of age. The results revealed that at light stimulus intensities of 0.01 and 3.0 cd/(s·m²), while the a-wave in heterozygous knock-in (Rho^P347S/+ ) mice showed no difference compared to the WT mice, there was a significant disparity in the b-wave between heterozygous knock-in (Rho^P347S/+ ) and homozygous knock-in (Rho^P347S/P347S ) mice (Fig. 4c, d). Intriguingly, further analysis of the ERG and HE staining results for Rho^P347S/P347S mice indicated a rapid rate of retinal degeneration, suggesting that they may not be ideal candidates as therapeutic model organisms (Fig. 4c–e).

Further ERG analyses of 2.5-month Rho^P347S/+ model mice compared to WT mice revealed growing disparities in the a-wave and b-wave under varying light intensities. Notably, the difference became especially pronounced at a light stimulus intensity of 3.0 (Fig. 4g). In summary, these findings confirm the successful establishment of RHO-P347S disease model mice, exhibiting significant visual function differences compared to WT mice. Their phenotype mirrors the human condition induced by the P347S point mutation causing RP, aligning roughly with our initial expectations. Thus, the Rho^P347S/+ disease model mice can be used as a experimental group for our treatment strategies.

Knockout replacement-based gene therapy ameliorates defects of Rho^P347S/+ mice

We subsequently assessed the therapeutic efficacy of our strategy in the Rho^P347S/+ model. The ERG waveforms of the treatment group were notably superior under various light stimulations compared to the AAV-GFP group at 3 months after injection (Fig. 5a). Moreover, analysis of extensive ERG data revealed a significant difference in a-wave and b-wave of the treatment group relative to the control groups under a 3.0-intensity light stimulus. Histologically, there was a notable increase in the number of surviving photoreceptors (Fig. 5c, d). A comparative study of retinal sections stained with HE between the treated and control groups showed that the treatment group had a thicker outer segment layer (Fig. 5c, d). These observations indicate that our strategy successfully delays photoreceptor death and retards the retina degeneration in the Rho^P347S/+ disease model.

Figure 5.

The therapeutic effect of dual AAV strategy on Rho^P347S/+ mice. (a) Representative ERG waveforms from Rho^P347S/+ mice 3-month postinjection. Treatment-AAV (Treatment group) and control-AAV (AAV-GFP group) were injected into the subretinal space bilaterally in Rho^P347S/+ mice on P14. Left panel, light intensity = 0.01 cd/(s·m²); right panel, light intensity = 3.0 cd/(s·m²). (b) Quantification of dark-adapted a-wave and b-wave amplitude in treatment AAV-injected and control AAV-injected Rho^P347S/+ mice, 3-month postinjection. The error bars indicate the SD. AAV-GFP group n = 16, Treatment group n = 19. **p < 0.01; ****p < 0.0001; Two-way paired ANOVA analysis. (c) H&E staining of Rho^P347S/+ mice 3-month postinjection. INL, ONL, IS/OS. Scale bar, 20 μm. (d) Counting of ONL cells per 100 μm. The number of cells on ONL layer at random 100 μm sites was counted and recorded. The error bars indicate the SD. n = 6; ***p < 0.001. Paired Student's t-test. (e) Thickness measurement of IS/OS layer. The error bars indicate the SD. n = 8; ***p < 0.001. Paired Student's t-test. IS/OS, inner segment/outer segment.

Discussion

In this study, we effectively demonstrated the excellent editing capabilities of the novel editing tool, CRISPR/Cas12i^HiFi, in combination with AAV in vivo, and coupled with the complementation of the human RHO gene, we successfully decelerated the disease progression in P23H and P347S RHO-point mutation disease models. Our findings not only pave the way for the application of Cas12i in gene therapy but also offer a promising alternative approach for RHO-mediated adRP disease.

For adRP caused by RHO point mutations, therapeutic strategies primarily fall into two categories: (1) mutation-specific strategy, which utilizes gene-editing tools for allelic-dependent editing at various mutation loci,^15,16 and (2) mutation-independent strategy, where the suppression or knockdown of endogenous Rho gene is coupled with the introduction of a correct RHO gene for treatment.^17

–20 Our proposed approach falls under the mutation-independent category. Previous studies have reported the successful knockout of endogenous mut-Rho using SpCas9 followed by reintroduction of the correct Rho gene, rescuing retinal abnormalities in a mouse model hRHO^C110R .²² EDITAS Medicine has also reported progress in the RHO-adRP “knockout and replace” therapeutic strategy based on AAV-CRISPR-SaCas9 in murine and nonhuman primate models.³⁴ These studies underscore the viability of the “knockout and replacement” strategy and offer valuable insights for our research.

Cas12i is a variant of class V Cas proteins.^35,36 Considering the heightened sensitivity of class V Cas proteins to base mismatches, they are generally believed to have higher specificity than class II Cas9 proteins, which has been validated in numerous studies.^35,37,38 Cas12i, a relatively new editing tool, due to its different PAM restrictions, owns a broader editing range compared to Cas9. Although initially Cas12i did not exhibit the ability to edit genes in mammalian cells, continuous optimization by researchers has elevated its editing efficiency in mammalian cells.^24,38,39 Existing research suggests that Cas12i, in terms of gene therapy, has higher safety and a more expansive application range than Cas9.

Furthermore, the smaller size of Cas12i provides us with more flexibility for optimizing our AAV therapeutic vector. Given the sensitivity of photoreceptor cells to RHO expression, both overexpression and less expression can be detrimental. Prioritizing safety and efficiency, we used a highly specific promoter in our experiments, which was validated in Supplementary Fig. S1d. Ultimately, this study used Cas12i as the editing tool in conjunction with AAV, achieving effective clearance of the endogenous gene in disease model mice, thereby validating the in vivo gene-editing capability of Cas12i (Fig. 1d, f, g). This broadens the application scope of the innovative Cas12i gene-editing tool in gene therapy.

While numerous mutation sites are implicated in RHO-mediated adRP, most studies focused on positions aa23 and aa347.²⁹ Transgenic mouse models Tg-hRho^P23H and Tg-hRho^P347S , embodying these mutations, were established in the last century and have been frequently referenced in gene therapy literature.^10,11 Although these human gene-inserted models reflect the disease progression in humans, the coexpression of human and murine genes can cause discrepancies. This has spurred debates on whether they accurately mimic the human disease and its underlying mechanisms.³⁰ In this study, we constructed a mouse model with the native P347S point mutation to better understand the pathology, refined therapeutic strategies, and evaluated the universality of our treatment approach.

Both homozygous and heterozygous Rho^P347S mice showed disease progressions akin to the human condition.⁴⁰ The rapid degeneration observed in homozygotes aligns with previous studies on the transgenic mouse model carrying the human P347S gene mutation. The RHO-P347S mutation led to the production of toxic proteins, and an increased presence of the erroneous gene corresponds to accelerated cellular damage.¹⁰ These findings further validate the accuracy and success of our disease model construction.

In conclusion, our findings demonstrate that engineered CRISPR-Cas12i^HiFi-mediated targeting of the entire Rho gene, coupled with precise RHO reintroduction, can mitigate RP progression caused by mutations at different RHO loci. Our results present a promising alternative therapeutic strategy. However, there remain numerous challenges that we need to address. Long-term examination of adRP mice treated by Cas12i^HiFi should be explored in the future to check potential off-target effects. In addition, further engineering of high-fidelity class V Cas system would also be needed to make this treatment safer and more effective.

Footnotes

Acknowledgments

We thank Shi-Wen Li and Xi-Li Zhu for their help with confocal laser scanning microscope; we appreciate Qing Meng's assistance in the operation of the Flow-Cytometer.

Authors' Contributions

W.L. and Q.Z. conceived the project and designed the experiments; X.S. and C.L. performed the experiments; Y.C. provided the Cas12i^HiFi plasmids and sequence; X.S., C.L., and T.C. analyzed the data; W.L., Y.Z., X.S., and C.L. wrote the article with the assistance of the other authors.

Author Disclosure

No competing financial interests exist.

Funding Information

This study was supported by grants from the National Key Research and Development Program (2019YFA0110800 to Wei Li, 2019YFA0903800), the National Natural Science Foundation of China (32225030 to Wei Li), the CAS Project for Young Scientists in Basic Research (YSBR-012 to Wei Li), the China Postdoctoral Science Foundation (2022 M713133 to Tongtong Cui).

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Table S1

Supplementary Table S2

References

Verbakel

, van Huet

RAC

, Boon

CJF

, et al. Non-syndromic retinitis pigmentosa. Prog Retin Eye Res, 2018; 66:157–186.

Dias

, Joo

, Kemp

, et al. Molecular genetics and emerging therapies for retinitis pigmentosa: Basic research and clinical perspectives. Prog Retin Eye Res, 2018; 63:107–131.

Hartong

, Berson

, Dryja

. Retinitis pigmentosa. Lancet, 2006; 368(9549):1795–1809.

Daiger

, Sullivan

, Bowne

. Genes and mutations causing retinitis pigmentosa. Clin Genet, 2013; 84(2):132–141.

Jacobson

, Kemp

, Sung

, et al. Retinal function and rhodopsin levels in autosomal dominant retinitis pigmentosa with rhodopsin mutations. Am J Ophthalmol, 1991; 112(3):256–271.

Athanasiou

, Aguila

, Bellingham

, et al. The molecular and cellular basis of rhodopsin retinitis pigmentosa reveals potential strategies for therapy. Prog Retin Eye Res, 2018; 62:1–23.

Palczewski

. G protein-coupled receptor rhodopsin. Annu Rev Biochem, 2006; 75:743–767.

Nathans

, Hogness

. Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc Natl Acad Sci U S A, 1984; 81(15):4851–4855.

Nathans

, Piantanida

, Eddy

, et al. Molecular genetics of inherited variation in human color vision. Science, 1986; 232(4747):203–210.

10.

, Snyder

, Olsson

, et al. Transgenic mice carrying the dominant rhodopsin mutation P347S: Evidence for defective vectorial transport of rhodopsin to the outer segments. Proc Natl Acad Sci U S A, 1996; 93(24):14176–14181.

11.

Olsson

, Gordon

, Pawlyk

, et al. Transgenic mice with a rhodopsin mutation (Pro23His): A mouse model of autosomal dominant retinitis pigmentosa. Neuron, 1992; 9(5):815–830.

12.

Machida

, Kondo

, Jamison

, et al. P23H rhodopsin transgenic rat: correlation of retinal function with histopathology. Invest Ophthalmol Vis Sci, 2000; 41(10):3200–3209.

13.

Yan

, Yao

, Li

, et al. Treatment of autosomal dominant retinitis pigmentosa caused by RHO-P23H mutation with high-fidelity Cas13X in mice. Mol Ther Nucleic Acids, 2023; 33:750–761.

14.

LaVail

, Yasumura

, Matthes

, et al. Ribozyme rescue of photoreceptor cells in P23H transgenic rats: Long-term survival and late-stage therapy. Proc Natl Acad Sci U S A, 2000; 97(21):11488–11493.

15.

Giannelli

, Luoni

, Castoldi

, et al. Cas9/sgRNA selective targeting of the P23H Rhodopsin mutant allele for treating retinitis pigmentosa by intravitreal AAV9.PHP.B-based delivery. Hum Mol Genet, 2018; 27(5):761–779.

16.

, Kleinstiver

, Leon

, et al. Allele-specific CRISPR-Cas9 genome editing of the single-base P23H mutation for Rhodopsin-associated dominant retinitis pigmentosa. CRISPR J, 2018; 1(1):55–64.

17.

Cideciyan

, Sudharsan

, Dufour

, et al. Mutation-independent rhodopsin gene therapy by knockdown and replacement with a single AAV vector. Proc Natl Acad Sci U S A, 2018; 115(36):E8547–E8556.

18.

Orlans

, McClements

, Barnard

, et al. Mirtron-mediated RNA knockdown/replacement therapy for the treatment of dominant retinitis pigmentosa. Nat Commun, 2021; 12(1):4934.

19.

Tsai

, Wu

, Lee

, et al. CRISPR-based genome surgery for the treatment of autosomal dominant retinitis pigmentosa. Ophthalmology, 2018. 125(9):1421–1430.

20.

Millington-Ward

, Chadderton

, O'Reilly

, et al. Suppression and replacement gene therapy for autosomal dominant disease in a murine model of dominant retinitis pigmentosa. Mol Ther, 2019; 19(4):642–649.

21.

Millington-Ward

. Strategems in vitro for gene therapies directed to dominant mutations. Hum Mol Genet, 1997; 6(9):1415–1426.

22.

, Tsai

, Huang

, et al. CRISPR genome surgery in a novel humanized model for autosomal dominant retinitis pigmentosa. Mol Ther, 2022; 30(4):1407–1420.

23.

Pacesa

, Lin

, Cléry

, et al. Structural basis for Cas9 off-target activity. Cell, 2022; 185(22):4067–4081.e21.

24.

Chen

, Hu

, Wang

, et al. Synergistic engineering of CRISPR-Cas nucleases enables robust mammalian genome editing. Innovation (Camb), 2022; 3(4):100264.

25.

Feodorova

, Koch

, Bultman

, et al. Quick and reliable method for retina dissociation and separation of rod photoreceptor perikarya from adult mice. MethodsX, 2015; 2:39–46.

26.

, Teng

, Li

, et al. Simultaneous generation and germline transmission of multiple gene mutations in rat using CRISPR-Cas systems. Nat Biotechnol, 2013; 31(8):684–686.

27.

Asokan

, Schaffer

, Samulski

. The AAV vector toolkit: Poised at the clinical crossroads. Mol Ther, 2012; 20(4):699–708.

28.

Dryja

, Hahn

, Cowley

, et al. Mutation spectrum of the rhodopsin gene among patients with autosomal dominant retinitis pigmentosa. Proc Natl Acad Sci U S A, 1991; 88(20): 9370–9374.

29.

Dryja

, McGee

, Hahn

, et al. Mutations within the rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. N Engl J Med, 1990; 323(19):1302–1307.

30.

Sakami

, Maeda

, Bereta

, et al. Probing mechanisms of photoreceptor degeneration in a new mouse model of the common form of autosomal dominant retinitis pigmentosa due to P23H opsin mutations. J Biol Chem, 2011; 286(12):10551–10567.

31.

Sakami

, Kolesnikov

, Kefalov

, et al. P23H opsin knock-in mice reveal a novel step in retinal rod disc morphogenesis. Hum Mol Genet, 2014; 23(7):1723–1741.

32.

Wang

, Yang

, Shivalila

, et al. One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell, 2013; 153(4):910–918.

33.

Richardson

, Ray

, DeWitt

, et al. Enhancing homology-directed genome editing by catalytically active and inactive CRISPR-Cas9 using asymmetric donor DNA. Nat Biotechnol, 2016; 34(3):339–344.

34.

Liu

, Wolf

, Dong

, et al. A mutation-independent CRISPR/Cas9-based ‘Knockout and Replace’ strategy to treat rhodopsin-associated autosomal dominant retinitis pigmentosa. Invest Ophthalmol Vis Sci, 2022; 63(7):3474.

35.

Yan

, Hunnewell

, Alfonse

, et al. Functionally diverse type V CRISPR-Cas systems. Science, 2019; 363(6422):88–91.

36.

Zhang

, Li

, Xiao

, et al. Mechanisms for target recognition and cleavage by the Cas12i RNA-guided endonuclease. Nat Struct Mol Biol,, 2020; 27(11):1069–1076.

37.

Kleinstiver

, Tsai

, Prew

, et al. Genome-wide specificities of CRISPR-Cas Cpf1 nucleases in human cells. Nat Biotechnol, 2016; 34(8):869–874.

38.

McGaw

, Garrity

, Munoz

, et al. Engineered Cas12i2 is a versatile high-efficiency platform for therapeutic genome editing. Nat Commun, 2022; 13(1):2833.

39.

Zhang

, Kong

, Xue

, et al. An engineered xCas12i with high activity, high specificity, and broad PAM range. Protein Cell, 2023; 14(7):538–543.

40.

Meng

, Ragi

, Tsang

. Therapy in rhodopsin-mediated autosomal dominant retinitis pigmentosa. Mol Ther, 2020; 28(10):2139–2149.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

1.46 MB

Knockout and Replacement Gene Surgery to Treat Rhodopsin-Mediated Autosomal Dominant Retinitis Pigmentosa

Abstract

Introduction

Materials and Methods

Animals

Construction of Cas12i HiFi plasmid and sgRNA

Subretinal injections

Production of AAV

Editing efficiency detection by T7EI

RNA extraction and RT

Real-time fluorescence quantitative PCR

Western blot

Immunofluorescence

Histological analysis

Retinal whole-mount

Electroretinography

Retinal dissociation and flow cytometry

Construction of RhoP347S knock-in mice

Intracytoplasmic RNA and DNA microinjection

Genomic DNA extraction, PCR, and sequencing for Rho P347S mice

Quantification and statistical analysis

Results

Dual AAV-Cas12i HiFi -based Rho knockout

Subretinal delivery of the dual AAV system efficiently coinfects photoreceptor cells

Long-term preservation of retinal function and structure with dual-AAV strategy (knockout+replacement)

Derivation of RhoP347S/+ knock-in mice

Knockout replacement-based gene therapy ameliorates defects of RhoP347S/+ mice

Discussion

Footnotes

Acknowledgments

Authors' Contributions

Author Disclosure

Funding Information

Supplementary Material

References

Supplementary Material

Construction of Cas12i^HiFi plasmid and sgRNA

Construction of Rho^P347S knock-in mice

Genomic DNA extraction, PCR, and sequencing for Rho ^P347S mice

Dual AAV-Cas12i^HiFi-based Rho knockout

Derivation of Rho^P347S/+ knock-in mice

Knockout replacement-based gene therapy ameliorates defects of Rho^P347S/+ mice